All Activity
This stream auto-updates
- Past hour
-
Is there a way to do the for loop inside vop,or I need to do the for loop outside the vop?
- Today
-
because Import Point Attribute VOP imports attribute value from the input geometry, not the modified one from previous iteration
-
Using the "Planar Patch From Curves" gives really nice results. Leaf_v0103.hiplc
- Yesterday
-
Any ideas on how to get a cratered moon-like surface?
smbell replied to smbell's topic in General Houdini Questions
Really cool, I'll have to check that out further. This was more just for me to learn heightfields. I just finished a maya project using actual lunar data using http://imbrium.mit.edu/EXTRAS/SLDEM2015/TILES/JP2/. -
Any ideas on how to get a cratered moon-like surface?
smbell replied to smbell's topic in General Houdini Questions
That's a pretty nice way of doing it! I appreciate your help. I tend to overthink things the longer I sit here and rack my brain over it. -
visualising attribs on packed geo points
richnosworthy replied to richnosworthy's topic in General Houdini Questions
Thanks guys, that makes sense finally. Cheers for the help -
visualising attribs on packed geo points
anim replied to richnosworthy's topic in General Houdini Questions
alternatively just append Repack SOP after native Assemble SOP and check Repack Packed Fragments -
visualising attribs on packed geo points
toadstorm replied to richnosworthy's topic in General Houdini Questions
The reason for this is a bit arcane if you're new to Houdini. The Assemble SOP creates a kind of packed geometry called "packed fragments," which means that rather than a typical packed primitive that is a reference to a single file or SOP data, it's a reference to a FRAGMENT of that data. This is more memory efficient when you have a single object that's been shattered into a bunch of pieces, rather than lots of copies of a few identical objects. The side effect of this is that the viewport shader that can tint packed primitives when you assign a color attribute to them won't work on packed fragments. If you wanted to tint these things individually in the viewport, you'd have to either tint them BEFORE you pack, or you'd have to make regular packed geometry instead of fragments. I'm attaching a file where I just cracked open that Assemble SOP and toggled the "Create Packed Fragments" option on the Pack SOP near the end of the chain. packedgeoattributes_nofragments.hiplc -
Any ideas on how to get a cratered moon-like surface?
Atom replied to smbell's topic in General Houdini Questions
You could use the actual NASA data. -
May be somthing like this ? It's not too bad except at the end of the leaf. Probably some tethraedralize node (wich I'm not comfortable with) settings. Hope that helps. Leaf_v0102 F.hip
-
Why the same operation using VEX and vop, the result is different? forloop.hiplc
-
Any ideas on how to get a cratered moon-like surface?
jonah@jonahtobias.com replied to smbell's topic in General Houdini Questions
hey smbell, here's a simple approach thats quick and might do what you want (not sure) cratered_terrain_12.hiplc -
Hey, i'm still pretty new to houdini but just trying to get my head around something (sorry if this is a stupid question). When you pack geometry, the geo gets saved as a reference and the packed geo represented by a primitive. But after its packed you can still add attributes to the new packed geometry points. For example i've seen people adding rotation and scale attributes that can then be used in a primitive sop to transform the packed geo. So say i added a randomised float attribute to the pieces after its been packed (maybe i'm going to use this in a wrangle to randomise the @P.y of the points later on), is it possible to visualise the attribute on the geo either with a visualise or color sop? So far i've not been able to do so. I know what i'm doing is probably wrong anyways, but just trying to understand why if the attribute is there on the packed point, why can it not be visualised on the geo like a tint on the geometry? Because the points are now locked away and can't have their color set right? Heres an example file of me creating a falloff attrib and trying to visualise that on the points Thanks guys packedgeoattributes.hiplc
-
FLIP sim moving POP particles? (Bubbles emitted from diver)
anim replied to bentway23's topic in Effects
you can advect your particles in a number of ways using vel field from FLIP sim either use POP sim with POP advect by volumes, or Slover SOP with VDB Advect Points or custom VOP/VEX using Advect By Volume VOP or volumesamplev() to directly sample the velocity and do your own advection alternatively you can isolate subset of particles by 'id' from the flip sim and instance your bubbles or other things to them -
I'm doing a FLIP sim with animated characters matchmoved to footage that was shot at 150 fps, so I'm trying to do a sim with pre-slowed actors. I've adjusted the time scale to 23.976/150, which puts it in the ballpark, but it still feels a little not quite right to me. Is this the approach I should take, adjusting the solver's time scale, or is there another setting elsewhere that would be better?
-
I'm looking to find a way to make a FLIP fluid sim move particles, which would themselves be used to instance underwater particulate matter or bubbles. All of my Googling turns up ways to make POPs influence FLIPs, but I'm unable to find this, and I'm sure it's depressingly easy. The scene is a pre-animated character diving into water. To get the bubbles as he dives it looks like the best bet will be to emit particles with instanced geo bubbles (correct? No matter how splashy I make the sim, I'm not getting any bubbles happening from the dive itself). Thanks for any help!
-
Any ideas on how to get a cratered moon-like surface?
smbell posted a topic in General Houdini Questions
I'm trying to achieve a moon surface with craters that can overlap at least semi correctly and have crater inside crater action. It's going to be eroded and processed after, so it doesn't' have to be totally perfect. I've tried a for- loop with height field layer, but I can't get them to replace each other and overlap correctly, I've also tried a for-loop with feedback with attribute transfer and a vops ramp, but same deal, I can't get them to layer correctly. I played with cops for a while, with a loop in there, but it seems like there has to be a solution using geometry. I've spent a couple days on this, so I'm officially looking for some advice. cratered_terrain_11.hiplc Edit: I haven't had a chance to try it, but maybe I could flatten the terrain in a step before layering in the new crater so it would always be layered over nothing... -
I don't want to sound mean or discouraging, but your network is a mess. It really does look like you have spent a lot of frustrating time working over and over it. I sympathize, I truly do. I've been there, many times. In essence you're trying to solve several problems at the same time. The easiest, quickest fix is to start with a simpler setup(s) to understand the essence of a constraint networks. In the attached file I've included a couple standard help scenes (I love the scenes included in the Houdini help docs and so should you) that should help you along. And I've included a quick custom setup showing the rbdmaterial fracture generated constraints. Do not use packed, you have to see your constraints previewed in red or they're not there, name your glue constraints "Glue" or they won't work. Best of luck! constraintHelp.hip
-
I'm trying to animate this tank tread but I'm really a newbie in animation can you help me ? thx hdni_tread1.hipnc
-
I was playing around with an RBD domino setup and extended it to accept a graphics based image (i.e. logo) for input. The Trace node converts the image to the outline of the logo, which is connected to a CopyToPoints, to setup the dominoes. ap_logo_dominoes_071819.hiplc
-
Actually its not working for what I need. I need to have the exactly same edges, no more then what i have in the lines thing. Just like this (but this one is fake) so I can select the primitives individually to do some things
-
Nice, thx Atom (once again!) I like the one with the divide better since I need to keep those base lines
-
Sorry about the late reply, didn't see this thread. You can send me MOPs questions directly, or ask on the MOPs forum. Relaxation is based on a simple spherical object radius, so if your objects aren't roughly sphere shaped you might get some overlaps. I'm not sure what exactly "didn't work" means in this context, but if you can provide a more detailed description, or better yet a HIP file, I could help you out. As MOPS instances are just packed primitives, the other thing you could try is to run them through a quick RBD simulation. Give the instances the point attribute `i@found_overlap=1` and then run them through a rigid body simulation, and they should separate.
-
Triangulate2D node can get you started, then you can play around with remeshing or poly reduce. ap_Leaf_v0102.hiplc
-
hi,thanks for checking my file. The thing is when I write the same function in VEX, the result is more visible in 30 loop counts. As for the vop solution, it seems it stopped functioning after several iterations. I will put new file here when I get home.
